Smart Mobs

by Howard Rheingold

  He makes his position clear regarding some of the fundamental research that Media Lab continues to pursue:

  The smart room is a retrograde concept that empowers the structure over the individual, imbuing our houses, streets and public spaces with the right to constantly observe and monitor us for the purported benefit of ensuring we are never uncomfortable or forced to get up from the armchair to switch on a lamp. Is it any wonder that I first encountered hostility with regard to the direction my research was to take? The very essence of my research was antithetical to the corporate-approved smart room concept. WearComp contests the utility of ubiquitous computing, forcing us to reconsider the current research trend toward cameras and microphones everywhere in the environment watching and listening to us in order to be “helpful.”. . . My opposite approach clashed dramatically with MIT’s research thrust—a corporate-directed set of priorities that privileges things over people.62

  Cyborg communities in what Mann calls “cyborgspace” are not, in his view, a dehumanized dystopia but are a defensive strategy against technological tyranny. Mann proposes that citizens can protect themselves against new concentrations of surveillance and control made possible by these technologies only by widely using the self-controlled, technically privatized form of wearable computer. The characteristic of sensory encapsulation is important to Mann. As he navigates the world, he sees only the output of his head-mounted cameras, as they are filtered through his wearable computer. Mann can turn the visual background of the world surrounding him into black and white and make his study materials pop up in color when he wants to study in a public place. Mann’s reaction to the technologically enhanced Society of the Spectacle that surrounds him is to use WearComp to filter commercial advertisements out of his visual field. The words and pictures on billboards become invisible at his command. Mann concedes that technology has made it possible for commercial interests to bombard modern urbanites with unsolicited sights and sounds designed to nudge them toward consuming a product or service. Mann’s radical social proposal is that the encapsulation of humankind in unsolicited commercial messages can only be reversed by using personal mediated-reality technology to filter our informational input and output.

  Mann understood that one cyborg would not change society. True democratization would only grow from mass adoption of the technology. The community of cyborgs he had longed for in the lonely early decades of his life slowly began to emerge. A documentary film about Mann, titled Cyberman, premiered in March 2002.63

  Another MIT Media Lab student, Thad Starner, was already engaged in wearable computer research when Mann arrived at MIT. Starner’s Ph.D. thesis chronicles his experience of using state-of-the-art wearable computers to remain in touch with his personal communications, the entire Web, and other cyborgs wirelessly linked together into a communal cyborgspace. Starner, now assistant professor and director of the Contextual Computing Group at Georgia Institute of Technology’s College of Computing, has worn his wearable computer continually since 1993.64 Although Mann’s online presence attracted the most attention, Starner deserves credit as one of the founders of wearable computing. In his thesis, he described the experience of reading email while walking through the halls at MIT. As others started wearing prototype wearable computers, “cyborgspace” started to become a site of social interaction.

  Starner reported how he and another cyborg realized that their face-to-face conversations were punctuated by natural breaks that noncyborgs might not understand; each computer wearer in the conversation waited periodically while the other paused to take a note or look up something online in the middle of the dialogue. Starner examined his own behavior while brainstorming alone and realized that he used the computer’s memory to “hold his place” on certain thoughts. He offered an example of something that happened to him in class:

  “What did we say was the importance of deixis?” asked the lecturer. With the end of the term approaching, the class was reviewing their study of discourse analysis.

  Volunteering, I said, “We said the importance of deixis is . . . uh . . . uh . . . humph, whoops! Uh, I’ll get back to you on that.”

  The class, most of whom were Media Laboratory graduate students familiar with wearable computing, began to laugh. I had not known the precise wording of the answer and had tried to retrieve my class notes on the topic. Having done this routinely in the past, I had expected to have the information in time to complete my sentence. Due to a complex series of mistaken keystrokes, I had failed so badly that I could not cover my error, much to everyone’s amusement.

  One of the members of the class leaned over and said, “You actually do that sort of thing all the time, don’t you? Now I’m impressed.”

  The combination of computer messaging tools, wireless connectivity, and a head-up display make such situations possible. In fact, members of the MIT wearable computing community and their colleagues take such an ability for granted. This informal networking can be used to encourage social gatherings, as above. It can also be used to form a type of “intellectual collective.”65

  Starner demonstrated the power of the collective by challenging reporters to ask him questions. Tharner used his handheld keyboard and wireless Net connection to send the question to all the computer users, mobile or stationary, who subscribe to the intellectual collective known as “the help instance.” Students doing homework in their dorm room or the library, computer wearers walking across campus, can see the message and respond to it if they know the answer or have access to the appropriate reference.

  Mann views the current fashionability of WearComp skeptically: “We cannot assume that all wearable technologies will be empowering. The ease with which researchers, sniffing the winds of technological change, switched from smart rooms to smart clothes (all the while maintaining what can only be described as a corporate ideology) clearly indicates the danger of blanket assumptions concerning the benefits of wearable technology. One can envision many wearable systems that, unfortunately, will take us in the other direction—away from personal freedom.”66 Mann cautions against WearComp that is sold as an individually empowering technology, but which is actually a “double agent” for some other institution or enterprise who wants to control or influence people.67 The technical question with politically important implications is, Who controls the information that comes into the WearComp and radiates out from the WearComp to sentient devices in the world?

  Since Mann left to inspire new generations of wearable computer students at Toronto and Starner started incubating wearable computer research at Georgia Tech, wearable computing has grown to be a major focus at Media Lab. A group at MIT is creating MIThril, the next generation of wearable computer, named after a magical garment in Lord of the Rings: “Our goal is to not simply build a platform, but to build a community of researchers, designers, and users.”68

  Perhaps more important than state-of-the-art wearable computer research platforms is the emergence of a hobbyist community similar to the subcultures that preceded and drove the PC and the Web. It didn’t take long to find out that cyberspace is full of Web sites and mailing lists devoted to wearable computers. That’s how I came to meet a cyborg in the lobby of New York’s Roosevelt Hotel. The president of Pakistan was staying at the same hotel that night, so the place was full of Secret Service agents, but Melanie McGee was sitting at a table in black leather and much more: head-mounted display over one eye, wearable computer in a shoulder holster, electrical cables wrapped in black electrical tape linking them, and battery pack on her belt. She composed email with one hand, held a drink in the other. Young men standing around the hotel lobby glanced at her and then talked into their sleeves. She’s a programmer and independent software developer when she’s not plugging components into her wearable computer. Even though only one of us wore a wearable computer as we walked around Manhattan, I found that the conversation could be augmented by easy access to Google. (Imagine conversations in the days before you could look up the answer
to any question.) Melanie is an example of the first wave of enthusiasts, those technically savvy enough to roll their own. The mass market, however, needs a well-designed and affordable plug-and-play system that requires little tinkering.

  IBM, Hitachi, and startups are vying to provide the equivalent of the Apple II that kicks the enthusiast community to the next level. Industry analyst Gartner Consulting predicts that 40 percent of adults and 75 percent of teenagers will use wearable computing devices by 2010.69 Xybernaut sells a voice-activated wearable computer to customers including Bell Canada. The Xybernaut unit, weighing less than two pounds, includes a head-mounted color display, microphone, and optional video camera.70 IBM and Citizen Watch Company announced WatchPad, a wrist-wearable computer with Bluetooth (a short-range radio technology) and infrared connectivity, speaker, microphone, video display, and fingerprint sensor.71 In January 2002, Hitachi announced the rollout of their Wearable Internet Appliance, combining a head-mounted display and eleven-ounce computer, running Microsoft’s Windows CE operating system, to be supplied to Xybernaut. Initial price for the system was around $2,200.72 In February 2002, Timex started test-marketing a new watch equipped with a radio frequency transponder that can be linked to a credit or debit card, enabling wearers to pay instantly at Exxon and Mobil gas stations and over 400 McDonald’s restaurants in the Chicago metropolitan area—by waving their arms.73

  Whether it comes through penny chips, wearable computers, geo-coded handheld devices, location-based services, smart rooms, digital cities, or sentient furniture, it seems clear that the next ten years will see more inanimate objects joining the Web and more people linked through mobile group-forming network technologies. The power of individuals to use smart mob media to form beneficial ad-hocracies—the power to solve social dilemmas—depends less on computing power or communication bandwidth and more on trust and willingness to risk the sucker’s payoff. That’s where reputation could make a crucial difference.

  5

  The Evolution of Reputation

  I used to live off reputation servers. Let’s say you’re in the Regulators— they’re a mob that’s very big around here. You show up at a Regulator camp with a trust rep in the high nineties, people will make it their business to look after you. Because they know for a fact that you’re a good guy to have around. You’re polite, you don’t rob stuff, they can trust you with their kids, their cars, whatever they got. You’re a certified good neighbor. You always pitch in. You always do people favors. You never sell out the gang. It’s a network gift economy.

  —Bruce Sterling, Distraction, a Novel

  Cooperation Catalysts

  You walk into a jewelry store, select a wristwatch, and hand a plastic card to the proprietor, who slides the card through a small device and waits a few seconds. You hear the “zip, zip, zip” of a cheap printer. The jeweler accepts your signature and trades you a valuable timepiece for whatever your card disclosed to the jeweler’s device and whatever you scribbled on the receipt. You can say the right numbers into a telephone or if you have an Internet connection and a good credit history, you can buy the same watch by typing the right numbers on your computer keyboard. Parts of the smart mob infrastructure are in place now, merely lacking a series of software upgrades already underway. Online credit verification services that have worked for decades are an ideal carrier for more finely nuanced reputation repositories capable of forecasting your taste in music, vouching for the trustworthiness of your computer code, attesting to your ability to evaluate wines, as well as verifying your credit record.

  Reputation marks the spot where technology and cooperation converge. The most long-lasting social effects of technology always go beyond the quantitative efficiency of doing old things more quickly or more cheaply. The most profoundly transformative potential of connecting human social proclivities to the efficiency of information technologies is the chance to do new things together, the potential for cooperating on scales and in ways never before possible. Limiting factors in the growth of human social arrangements have always been overcome by the ability to cooperate on larger scales: the emergence of agriculture ten thousand years ago, the origin of the alphabet five thousand years ago, the development of science, the nation-state, the telegraph in recent centuries, did more than accelerate the pace of life and make it possible for the human population to expand. These cultural levers also enlarged the scale of cooperation, radically altering the way people live.

  The totalitarian threat posed by the prospect of mobile and pervasive media demands attention. The possibility that breakouts of cooperation could expand liberty is also worthy of attention.

  Consider how early versions of “reputation management” or “social filtering” systems currently support new forms of broad-based cooperation:

  Electronic communication networks transformed the centuries-old institutionalized trust system of banking. Today’s global institutionalized trust system of credit cards and ATMs, backed up by instantaneously available credit databases, authenticates millions of financial transactions every day.

  eBay, dominant survivor of the e-commerce bubble, uses a reputation system to facilitate billions of dollars worth of transactions for people who don’t know each other and who live in different parts of the world.

  Epinions pays contributors of the most popular online reviews of books, movies, appliances, restaurants, and thousands of other items. Epinions’s reputation system enables people to rate reviewers and to rate other raters through “webs of trust.” The most trusted reviewers are read by more people and therefore make more money.

  Slashdot and other self-organized online forums enable participants to rate the postings of other participants in discussions, causing the best writing to rise in prominence and most objectionable postings to sink.

  Amazon’s online recommendation system tells customers about books and records bought by people whose tastes are similar to their own.

  Google.com, the foremost Internet search engine, lists first those Web sites that have the most links pointing to them—an implicit form of recommendation system.

  Hordes of programmers who compete for bragging rights as well as paying work are already driving the evolution of the first-generation reputation systems toward more advanced forms. Upendra Shardanand and Pattie Maes at the MIT Media Lab started something growing on the Net when they introduced Ringo, the “social information filtering” system that recommended music on the basis of shared tastes.1 The MIT researchers “automated word-of-mouth recommendations” with computational methods. Users were invited to send an email command to the Ringo server. Return email presented a list of 125 musicians. Each user rated the musicians he or she liked. Ringo performed statistical correlations and then suggested new artists to each person—musicians they might not have known about but who were recommended by people with similar musical preferences.

  Ringo launched in July 1994 and grew to more than 2,000 users by September. The MIT researchers started a company named Firefly to commercialize Ringo and sold it to Microsoft in 1998. Microsoft eventually implemented its own version of Firefly’s “digital passport” technology.2 Ringo turned out to be the progenitor of an evolutionary lineage.

  Finding new books, movies, or music is a popular pursuit, but it represents only one form of the myriad webs of trust that support markets, scientific enterprises, businesses, and communities. Consider the history of online knowledge-sharing economies. One of the most seductive aspects of social cyberspace is the way virtual communities share useful information. I remember how excited I became in the 1980s, when the never-ending “Experts on the Well” discussion inspired people in the Well, a virtual community of a few thousand, to compete for the honor of providing the fastest and most accurate answers online to questions posed by other members of the community.3 This custom is more sophisticated than automated word-of-mouth systems like Ringo because it requires each human recommender to keep in mind many other people’s intellectual preferences, glean
ed solely from online conversations.

  Trading know-how isn’t new. Developing a reputation for distributing high-quality recommendations is one way to accrue social status, and humans have extraordinary talents for social games. Trading know-how with people on six continents in real time, however, is more than just new; it fundamentally transforms knowledge-sharing by drastically lowering the transaction cost of matching questions and answers. While surfing the Web, little extra effort is required to send email to friends with a URL pointing to an interesting page, and little effort is required in finding the right specialized forum to ask a question. More recently, the phenomenon of “weblogging,” which enables thousands of Web surfers to publish and update their own lists of favorite Web sites, has tipped online recommendation-sharing into an epidemic.

  I stated something long known to Usenet veterans when I wrote in 1988 that one of the most attractive social innovations enabled by virtual communities was the way members could “serve as information hunters and gatherers for each other.”4 Four years later, researchers at Xerox PARC applied a more systematic version of a hunting-and-sharing methodology, introducing the term “collaborative filtering” in 1992 to describe their Information Tapestry. Information Tapestry software enabled researchers to annotate documents as they read them and made it easy for other researchers to use recommendations to find useful documents.5

  Informal social aggregation of useful knowledge goes back to the lists of frequently asked questions (FAQs) posted to some Usenet newsgroups since the 1980s; these lists of questions and answers, accumulated through years of archived online conversations, were compiled to prevent newcomers from besieging more knowledgeable posters with questions that had already been answered. Telling a newbie to “read the FAQ!” is a way for a group to constrain the overconsumption of a public good; experts contribute knowledge as long as the conversation retains their interest but stop contributing if new-bies’ questions dominate the conversation. Beyond their defensive function, FAQs constitute a new kind of encyclopedia in themselves, collectively gathered, verified, articulated webs of knowledge about hundreds of topics.6